Multi-media trancoding includes video transcoding, audio transcoding, image transcoding, and so on. Video transcoding is transferring a compressed coded video stream to another video stream, to ensure the video applying for different network bandwidths, different handling capacity of terminals, and different user's demands. Traditional way of video transcoding does not slice single video, but trancodes the video directly, or trancodes after slicing the video. Slicing means to slice video in time domain and space. For example, slice a video to several parts with a few frames a unit, or slice a frame image to several parts in space. A slicing boundary is produced after slicing the video, the sliced parts can be trancoded parallel, thereafter all the trancoded parts are pieced together to a whole video. Audio transcoding is to transcode after slicing the audio in time domain. While image transcoding is to trancode after slicing the image in space.
Nevertheless, the traditional way of slicing multi-media will cause distortion or vague in the slicing boundary after piece the transcoded parts together, which makes the effect of the multi-media worse.